Clinical outcomes of PFNA vs. long-stem hemiarthroplasty for unstable proximal femur fractures in older adults: a retrospective cohort study
Abstract
The optimal surgical management of unstable extracapsular proximal femur fractures (AO/OTA types A2 and A3) in elderly patients with osteoporosis remains controversial. This study compared clinical outcomes of proximal femoral nail antirotation (PFNA) vs. primary uncemented bipolar hemiarthroplasty with long femoral stem in patients aged ≥75 years. The primary outcome was reoperation rate within 24 months; secondary outcomes included perioperative parameters, functional recovery, complications and mortality. A retrospective comparative study was conducted including 174 patients (mean age 82.2 ± 7.1 years, 61.5% female) treated between January 2018 and December 2023. Patients underwent either PFNA ( n = 96) or hemiarthroplasty (HA, n = 78). Primary outcomes included perioperative parameters, functional recovery [Harris Hip Score (HHS), Barthel Index], complications, reoperations, and mortality at 12 and 24 months. Mean operative time was shorter in the PFNA group (59.6 vs. 74.2 min, P < 0.001), while HA patients achieved earlier ambulation (2.8 vs. 6.2 days, P < 0.001) and full weight-bearing (4.2 vs. 62.8 days, P < 0.001). HHS was higher in the HA group at 3 months (71.6 vs. 66.8, P < 0.001) and 6 months (78.2 vs. 74.6, P = 0.004), but similar at 24 months. Reoperation rates (7.7% vs. 19.8%, P = 0.03) and 2-year mortality (2.6% vs. 11.5%, P = 0.02) were significantly lower in the HA group. Multivariable analysis identified PFNA as an independent risk factor for complications (OR = 1.92) and reoperation (sHR = 2.78). In geriatric patients with unstable extracapsular femoral fractures, long-stem hemiarthroplasty was associated with faster early recovery, lower reoperation rates, and a lower observed 2-year mortality compared to PFNA, which may be considered as an alternative primary option in appropriately selected elderly patients. However, given the retrospective design and potential for residual confounding, this observed association should not be interpreted as a causal effect, which should be confirmed by future prospective randomized trials.